HEPro AI+ 🚀

AI-powered Decision Intelligence System for Proactive Student Mentoring

🚨 AI system that identifies at-risk students BEFORE failure and automatically assigns the right mentor using hybrid intelligence.

---

🎯 Problem Solved

Institutions lack systems that proactively identify struggling students and take action before failure occurs.

HEPro AI+ bridges this gap by combining behavioral analytics with automated decision-making.

---

🧩 What is HEPro AI+?

HEPro AI+ is an AI-powered decision intelligence system designed to understand students beyond marks.

Instead of waiting for students to fail, this system proactively identifies:

• hidden stress
• declining engagement
• productivity issues
• career confusion

…and automatically recommends what action should be taken — and who should take it.

It transforms raw student data into clear, actionable mentoring decisions.

---

⚡ What This System Actually Does

• Detects hidden high-risk students (even if grades look fine)
• Identifies behavioral patterns using ML clustering
• Assigns the most suitable mentor automatically
• Generates real-time intervention recommendations

➡️ Output: Actionable mentoring decisions, not just analysis ➡️ Designed to support early intervention and reduce student failure rates through proactive decision-making.

---

💡 Why This Project Matters

Most systems only show what is wrong. HEPro AI+ goes one step further — it answers:

“What should we do next?”

This shift from analysis → action is what makes it a Decision Intelligence System, not just an ML project.

This makes it not just a monitoring tool, but a decision-making engine.

---

🔄 How the System Works (Step-by-Step)

1️⃣ Data Generation

The system starts with a realistic dataset of students containing:

• academic performance
• attendance
• stress & wellness indicators
• productivity & focus metrics

---

2️⃣ Rule-Based Scoring (SRI)

Each student is evaluated using a composite score:

SRI (Student Readiness Index) based on:

• Academic Performance (APS)
• Wellness (WWS)
• Productivity (PTMS)
• Career Readiness (CRS)

This ensures interpretability and transparency.

---

3️⃣ Machine Learning (Clustering)

Using K-Means (K=4), students are grouped into behavioral personas such as:

• High-Achieving but Stressed
• Career Confused
• Stable & Balanced
• High-Risk Disengaged

This reveals behavioral patterns that traditional scoring completely misses.

---

4️⃣ Decision Intelligence Engine

This is the core of the system.

It decides:

• What is the student's biggest problem?
• What type of help is needed?
• Which mentor should be assigned?

Key logic:

• Wellness is always prioritized first
• High-risk students get immediate attention
• Mentor assignment is based on expertise + availability

---

5️⃣ Mentor Matching System

The system intelligently assigns:

• Career mentors
• Wellness counselors
• Productivity coaches

It also ensures:

• No mentor overload
• Balanced workload distribution

---

6️⃣ Outputs Generated

The system produces:

• 📄 final_recommendations.csv
  → Includes cluster label, assigned mentor, intervention type, and priority level

• ⚠ alert_log.txt
  → High-risk student alerts requiring immediate action

---

📊 Sample Outputs

These outputs demonstrate how HEPro AI+ transforms raw student data into actionable mentoring decisions with clear prioritization and intervention strategies.

🔹 Final Recommendations

🔹 High-Risk Alert Log

🔹 Cluster Visualization

---

📁 Project Structure

HEPro-AI-Plus/
│
├── assets/                    # Output screenshots
│   ├── final_recommendations_output.png
│   ├── alert.png
│   └── pca_cluster_visualization.png
|
├── data/                   # Input & processed datasets
│   ├── students.csv
│   ├── students_scored.csv
│   ├── students_clustered.csv
│   ├── mentors.csv
│   ├── mentors_assigned.csv
│   └── cluster_profiles.json
│
├── src/                    # Core system logic
│   ├── generate_data.py
│   ├── scoring_system.py
│   ├── run_clustering.py
│   └── run_matching.py
│
├── notebooks/              # Development & experimentation
│   ├── scoring_system.ipynb
│   ├── student_segmentation.ipynb
│   └── mentor_matching_system.ipynb
│
├── outputs/                # Final system outputs
│   ├── final_recommendations.csv
│   └── alert_log.txt
│
├── docs/                   # Supporting documentation
│   ├── SYSTEM_ARCHITECTURE.md
│   ├── SCORING_LOGIC.md
│   ├── CLUSTER_INSIGHTS.md
│   ├── DECISION_INTELLIGENCE_REPORT.md
│   └── MENTORING_GUIDE.md
│
├── requirements.txt
└── README.md

---

🚀 How to Run the Project

Step 1: Install Dependencies

pip install -r requirements.txt

---

Step 2: Run the Pipeline

python src/generate_data.py
python src/scoring_system.py
python src/run_clustering.py
python src/run_matching.py

---

📊 Data Flow

students.csv
   ↓
students_scored.csv
   ↓
students_clustered.csv
   ↓
final_recommendations.csv + alert_log.txt

⚡ Pipeline Snapshot

Data → Scoring → Clustering → Decision Engine → Mentor Assignment → Alerts

---

🧠 Key Features

• Hybrid AI Architecture (Rule-Based + Machine Learning)
• Explainable decision-making
• Early risk detection
• Smart mentor allocation
• Modular & scalable design

---

🚀 What Makes This Different?

• Not just prediction → decision-making system
• Combines rule-based logic + ML (hybrid AI)
• Prioritizes wellness over academics
• Includes mentor capacity constraints
• Fully explainable (no black-box decisions)

---

🔮 Future Improvements

• Feedback loop (learning from outcomes)
• Dashboard (Streamlit / Plotly)
• Real-time monitoring system
• Integration with institutional databases

---

💼 Use Case

This project demonstrates:

• Applied Machine Learning
• Decision Intelligence Systems
• Real-world problem solving
• Explainable AI design

Suitable for:

• AI/ML roles
• Data Science internships
• Backend + ML system design discussions

---

⚙️ Tech Stack

• Python
• Pandas, NumPy
• Scikit-learn (K-Means)
• Matplotlib / Seaborn
• Jupyter Notebook

---

📈 System Scale

• Simulated dataset of 50 students
• 12 mentors with capacity constraints
• End-to-end automated pipeline execution

---

👨‍💻 Author

Developer: Harshit Sharma | LinkedIn Profile

HEPro AI+ is not just analyzing students —
it is making the right decisions at the right time, for the right student.

---

⭐ Support

If you found this project useful or interesting, consider giving it a ⭐ — it helps increase visibility and motivates further development.